exam #2

0.0(0)
Studied by 0 people
call kaiCall Kai
learnLearn
examPractice Test
spaced repetitionSpaced Repetition
heart puzzleMatch
flashcardsFlashcards
GameKnowt Play
Card Sorting

1/119

encourage image

There's no tags or description

Looks like no tags are added yet.

120 Terms

1
New cards
  • true-experimental

  • quasiexperimental

  • pre-experimental


the traditional classifications of experimental (analytic) designs

2
New cards

true-experimental designs

the design type that is the criterion against which all other methodological approaches are judged

3
New cards

true-experimental

the experimental-type design that offers the greatest degree of control & internal validity

4
New cards

true-experimental design

identify the experimental-type design based on the notation

<p>identify the experimental-type design based on the notation</p>
5
New cards
  • randomization

  • control group

  • manipulation of an independent variable


the three major characteristics of true-experimental design

6
New cards
  • sample selection

  • group assignment


at what two points does randomization occur within the true-experimental design?

7
New cards

the extent to which conclusions can be generalized (external validity)

while a true-experimental design that does not use random sample selection is still considered true experimentation, it is limited in _______ .

8
New cards

randomization

powerful technique that is designed to increase control and eliminate bias by neutralizing the effects of extraneous influences on the outcome of a study

9
New cards
  • “attention factor”

  • “Hawthorne effect”

  • “halo effect”


terms that all refer to the phenomenon of the subject experiencing change as a result of simply participating in a research project

10
New cards

control group

allow investigators to observe the presence or absence of the “attention factor”/”Hawthorne effect”/”halo effect”

11
New cards

manipulation

the ability to provide and withhold the independent variable that is unique to the true experiment

12
New cards
  • posttest-only designs

  • Solomon four-group designs

  • factorial designs

  • counterbalance designs


four basic design variations of the true experiment

13
New cards

the absence of a pretest

the difference between classic true-experimental design and & posttest-only designs is:

14
New cards

posttest-only

identify the experimental-type design based on the notation

<p>identify the experimental-type design based on the notation</p>
15
New cards
  • pretesting is not possible/appropriate but the research purpose aims to examine causal relationships

  • the threat to learning is highly likely with repeated testing using the same measure of the dependent variable


when are posttest-only designs most valuable?

16
New cards

Soloman four-group designs

experimental-type design that allows investigators to evaluate the effect of testing on scores on the posttest & interaction between the test and the experimental condition

17
New cards

Solomon four-group design

experimental-type design that combines that true-experiment & posttest-only designs into one design structure

18
New cards

Solomon four-groups design

identify the experimental-type design based on the notation

<p>identify the experimental-type design based on the notation</p>
19
New cards

ability to detect interaction effects

the key benefit of the Solomon four-groups design

20
New cards

because it requires two additional groups, it is:

  • costly

  • time-consuming


why is the Solomon four-groups design infrequently used in health & human service inquiry?

21
New cards

factorial designs

research design that evaluates the effects of two or more independent variables OR the effects of an intervention on different factors/levels of a sample or study variables

22
New cards
  • descriptive

  • analytic


the two broad types of research designs

23
New cards

descriptive

research designs that are:

  • non-experimental

  • generate hypotheses


24
New cards

analytic

research designs that:

  • test (null) hypotheses

  • examine causal relationships


25
New cards

descriptive

in ______ research, researchers have a passive role.

26
New cards

analytic

in ______ research, researchers have an active role.

27
New cards

Participants

Outcome

Time

elements of PICOT focused on in descriptive research

28
New cards
  • the purpose/research question

  • available resources (money/time)

  • preferred exposure/outcome

  • ethical considerations


factors considered in the selection of a research design

29
New cards
  • control group

  • manipulation


which of the three true-experimental design elements are present in quasi-experimental designs?

30
New cards
  • nonequivalent control group

  • interrupted time series


the two basic types of quasi-experimental designs

31
New cards

pretest-posttest (pre-experimental) design

identify the experimental-type design based on the notation

<p>identify the experimental-type design based on the notation</p>
32
New cards

static group comparison

identify the experimental-type design based on the notation

<p>identify the experimental-type design based on the notation</p>
33
New cards

interrupted time series

research design that involves the repeated measurement of the dependent variable both before & after the introduction of the independent variable

34
New cards
  • investigator is searching for change over time

  • a causal attribution/claim is not the goal

  • random assignment is not appropriate, ethical, or feasible


in what three instances is quasi-experimentation most valuable?

35
New cards
  • answer level 1 (descriptive) or level 2 (relational) questions

  • generate pilot/exploratory evidence


pre-experimental designs can be of value to:

36
New cards
  • one-shot case study

  • pretest-posttest design

  • static group comparison

  • big data analytics


the four types of pre-experimental designs

37
New cards

one-shot case study

pre-experimental design type in which the independent variable is introduced; then the dependent variable is measured in only one group

38
New cards

manipulation

key characteristic of one-shot case (pre-experimental) studies

39
New cards

manipulation

key characteristic of pretest-posttest (pre-experimental) studies

40
New cards

manipulation

key characteristic of static group comparison (pre-experimental) studies

41
New cards

static group comparison

pre-experimental design in which a comparison group is added to the one-shot case study design

42
New cards

one-shot case study (pre-experimental)

midterm exams in a college course (meant to assess students’ learning) is most like what research design?

43
New cards

non-experimental designs

research designs that fit none of the three criteria for true experimentation (randomization, control group, manipulation) are present

44
New cards

when testing a concept/construct or set of relationships among constructs that naturally occur

when are non-experimental designs more useful?

45
New cards

post hoc, through statistical analysis

in non-experimental designs, any manipulation of the variables is done:

46
New cards
  • surveys

  • passive observation

  • ex post facto designs


the three non-experimental designs frequently used in health & human service research

47
New cards

survey designs

primarily used to measure characteristics or attributes of a population

48
New cards
  • describe population parameters/features

  • predict relationship among these phenomena


through survey designs, it is possible to:

49
New cards
  • investigators can reach a large # of respondents with relatively little expenditure & time

  • numerous variables can be measured by a single instrument

  • statistical manipulation of the data can permit multiple uses


the advantages of survey designs

50
New cards

correlational designs

passive observation designs are often referred to as:

51
New cards

after the fact

“ex post facto” means:

52
New cards

ex post facto designs

research designs in which the phenomena of interest have already occurred and cannot be manipulated in any way

53
New cards

ex post facto designs

research design frequently used to examine relationships between naturally occurring population parameters and specific variables

54
New cards

meta-analysis

research design in which research reports, rather than actual people, are surveyed as the unit of analysis

55
New cards
  1. specify topic

  2. specify type of finding of importance (outcomes or covariance)

  3. establish inclusion/exclusion criteria for studies

  4. identify, locate, & retrieve relevant studies

  5. establish a systematic approach to organizing research records

  6. derive numerical value or index representing each study’s effect size


the six basic steps in a meta-analysis

56
New cards

experimental-type meta-analysis

the purpose of _________ are to summarize, integrate, and interpret an empirical body of research or studies in which the outcomes are quantitative.

57
New cards

effect size

statistic that codes the magnitude of the effect (or outcome) as a result of being exposed to the independent variable

58
New cards

observational; experimental

analytic research designs may be _____ or _____ .

59
New cards

cohort

a well-defined group of individuals who share a common characteristic or experience

60
New cards
  • cross-sectional

  • cohort

    • prospective

    • retrospective

  • case-control


types of observational (analytic) research

61
New cards

cross-sectional

types of observational (analytic) research in which exposures & outcomes are collected at the same time

  • ~ “taking a picture”


62
New cards

1 (descriptive) & 2 (relational)

large datasets can explore what level(s) of experimental questions?

63
New cards
  • broad scope of data

  • eliminating need to collect data

  • data is collected by experts


benefits of using large datasets collected by organizations such as the CDC

64
New cards

prospective cohort study

type of observational (analytic) research design that is appropriate to examine data collected forward in time to compare outcomes

65
New cards
  1. define the sample & group allocation approach (self-selected based on behavior)

  2. consider the time frame & group-level changes

  3. compare outcomes between the exposed & unexposed


the three steps in a prospective cohort study

66
New cards

self-selection (behavior) → compared to a threshold set before the study begins

in a prospective cohort study, group membership is determined by:

67
New cards

risk ratio (highest incidence rate/lowest incidence rate)

how are the results of prospective cohort studies presented?

68
New cards
  • estimate prevalence rates

  • provide measure of risk

  • allow investigators to study rare exposures

  • low risk of recall bias

  • can assess multiple exposures and/or outcomes


advantages of prospective cohort studies

69
New cards
  • impractical for rare diseases

  • expensive & time-consuming

  • some diseases have long latency periods

  • is limited to risk factor @ baseline BUT behavior changes!

  • participant attrition (non-response, loss-to-follow-up)


disadvantages of prospective cohort studies

70
New cards

those exposed are ______ times as likely to report _____ .

how are risk ratios interpreted?

71
New cards

retrospective cohort study

type of observational (analytic) research design that is appropriate to examine data collected in the past

72
New cards
  1. identify time frame

  2. identify those exposed & unexposed to a risk factor

  3. compare outcomes between the two groups


the three steps in a retrospective cohort study

73
New cards

exposure staus

  • self-selected, NOT randomly allocated


in a retrospective cohort study, groups are separated by _______ .

74
New cards

case-control study

type of observational (analytic) research design that is appropriate to examine the influence of prior exposures and/or risk factors

75
New cards
  1. identify the time period

  2. understand group membership (cases vs controls)

  3. compare prior exposures/risk factors between groups


the three steps in a case-control study

76
New cards

disease status

in case-control studies, groups are separated by _______ .

77
New cards
  • proportion exposed

  • odds of outcome

  • odds ratio


how are the results of case-control studies presented?

78
New cards
  • proportion of cases exposed (cases exposed/all cases)

  • proportion of controls exposed (controls exposed/all controls)


how to calculate the proportion exposed, based on the results of a case-control study

79
New cards
  • for exposed: exposed cases/exposed controls

  • for unexposed: unexposed cases/unexposed controls


how to calculate the odds of outcome, based on the results of a case-control study

80
New cards

cross product:

(exposed cases)(unexposed controls)/(unexposed cases)(exposed controls)


how to calculate the odds ratio, based on the results of a case-control study

81
New cards
  • appropriate for studying rare events

  • can study multiple risk factors

  • can be performed quickly & at a relatively low cost


advantages of case-control studies

82
New cards
  • subject to recall bias

  • limited to studying one disease at a time

  • temporal relationship is unclear


disadvantages of case-control studies

83
New cards

Odds of ______ are similar between cases & controls.

how are the null hypotheses of case-control studies formatted?

84
New cards

one-shot case study

identify the experimental-type design based on the notation

<p>identify the experimental-type design based on the notation</p>
85
New cards

Mill’s canons

states 5 basic principles for inferring causality

86
New cards
  1. temporal sequence

  2. strength of association

  3. consistency

  4. dose response

  5. biological plausibility


the five basic principles for inferring causality, according to Mill’s canon

87
New cards

temporal sequence

identify the principle of Mill’s canons: stipulates that exposure to a risk factor must precede disease development

88
New cards

strength of association

identify the principle of Mill’s canons: states that large (& clinically significant) differences in disease must exist between exposed & unexposed individuals

89
New cards

consistency

identify the principle of Mill’s canons: states that an association is always observed between the presence of a risk factor & disease development

90
New cards

dose response

identify the principle of Mill’s canons: states that the risk of disease must be greater with stronger exposure

91
New cards

biological plausibility

identify the principle of Mill’s canons: states that the association between a risk factor & disease must be explainable by existing knowledge about biological mechanisms

92
New cards

if the study involves populations that are:

  • historically low in participation rates

  • may be likely to drop out mid-study


when may oversampling be necessary for a randomized control trial?

93
New cards

the variability of the outcome measure

a study’s sample size is dependent on:

94
New cards
  • establishes causality (internal validity)

  • reduces bias → distributes confounds


why is randomization crucial in experimental studies?

95
New cards
  • placebo

  • active

  • self


control types used in randomized control trials

96
New cards

concurrent (parallel)

the most common randomized control trial design

97
New cards

concurrent/parallel

identify the randomized control trial design: two (or more) groups are formed & studied simultaneously

98
New cards

waitlist (delayed treatment)

identify the randomized control trial design: two groups are randomly formed and:

  • one group receives the treatment

  • the control group later receives the treatment, if it turns out to be effective


99
New cards

wait-list (delayed treatment)

the randomized control trial design created to address the ethical concerns of not treating control groups

100
New cards

cross-over (counterbalanced)

identify the randomized control trial design: participants are randomly assigned to varying sequences of interventions